The primary objective of this study is to assess the role which prosody plays on the sentence comprehension processes of aphasic patients. Recent studies have shown that the anterior aphasic does not make normaluse of grammatical morphemes in deriving sentence structure. This finding has been cited as evidence of a specifically syntactic dificit. However, prosodically, grammatical morphemes are linguistically unstressed,and therefore of shorter duration acoutically than stressed content words. Other studies have shown that performance deficits in aphasia increase as the duraton of the stimulus decreases. Yet other studies show that normal subjects use sentence prosody as a cue to determining syntactic structure. The slaience of stressed vs. unstressed elements in linguistic and non linguistic strings, for aphasic plus non-aphasic groups will be assessed. In one part, this will be done by means of probe studies, in which subjects hear a string followed by a single element which they must judge as coming from the preceding string or not as quickly as possible. Stress will be manipulated, RT and error scores will be assessed. (2) in a second part, subjects willhear a linguistic or non-lingusitic string played repeatedly. They will be required to tap to a preselected word. The variance of tapping to stressed vs. unstressed elements will be assessed. Taken together, the studies will clarify our view at qyntactic deficits in aphasia, and relate these to more general, non-linguistic capacities.